What if a robot could do all those mundane work for you? Well, this could soon be a reality, as an Indian-American scientist is developing such intelligent robots which he claimed could help people with their everyday task.
Ashutosh Saxena, an assistant professor of computer science at Cornell University in the US, is working to bring robots into homes and offices that can clean up a messy room, assemble a flat-pack bookcase or unload a dishwasher, all without human intervention.
Saxena, who has done his B.Tech from IIT Kanpur in 2004 and joined the Cornell faculty in 2009, believes robots can make people's lives better and more productive, according to a
Cornell release.
"Just like people buy a car, I envision that in five to 10 years, people will buy an assistive robot that will be cheaper or about the same cost as a car," Saxena was quoted as
saying by a university release.
One of the biggest technical challenges, according to Saxena, is endowing robots with the ability to learn in uncertain environments.
It's one thing to make a robot do simple tasks such as picking up a pen or turning sides, but to make a robot understand how to pick up an unknown object or navigate an unknown room is quite different.
Saxena, who led the manipulation group in the STAIR project (Stanford Artificial Intelligence Robot) at Stanford University, has researched how to make robots perceive information in cluttered and unknown environments. His work has also enabled robots to estimate depth from a single image.
"For example, if you look at a new object, how would you pick it up? If you are in a new environment, how do you figure out how far away things are?" he said.
Saxena and his team have focused on how to make robots gather information in cluttered and unknown environments.
Using a camera, one of his robots can evaluate an object -- say a cup or plate and figure out how best to grab it.
This kind of technology will eventually become the basic capability of a full-fledged dishwasher-unloading robot, he said, hoping that such robots could be available within 10
years.
"In a cluttered room, it is notoriously difficult for today's object detection algorithms to reliably find an object as simple as a shoe," he said.
The key, he explained, is not to look at this task in isolation. If the three-dimensional structure of the room is known, it becomes easier to find the objects.
Researchers at the Upson Hall's Personal Robotics Lab, which Saxen heads, are building learning algorithms to enable roboticists to quickly combine several perception algorithms
into a more reliable one.
Their project has recently been presented at the European Conference on Computer Vision in Greece.